The recent swelling of the ranks of homeless has been attributed to a number of economic and social trends including the deinstitutionalization of psychiatric patients. Evidence from several surveys indicates that there are many psychiatrically impaired individuals among the homeless population. The provision of mental health and other services to the homeless presents major logistic and administrative problems. In order to plan for more effective service provision to this group, more extensive data are required to estimate their needs. In order to monitor the implementation of an intervention program based on these findings, a method must be developed for follow-up of this residentially unstable population. A survey of 600 homeless persons in the missions, shelters and jail in Baltimore City will be conducted. Pilot data will be gathered on the demographic composition of this population which will be used to create strata for sampling. Homeless persons will be interviewed by trained interviewers using psychiatric screening instruments (GHQ/MMSE) and both forced-choice and open-ended interview items on sociodemographic, health and life history characteristics. All persons who screen positive for psychiatric disorder and a proportion of those who screen negative will be selected for psychiatric evaluation and physical examination which will form the basis of a need for treatment statement each subject (N=300). The data gathered from the survey and examination will provide information concerning how persons become homeless including the role of deinstitutionalization, and allow estimation of needs for service among shelter-using homeless persons. About one-third of the examined group will be asked to participate in a six month follow-up study to test a methodology for monitoring homeless persons and to provide data on movement and patterns of utilization of shelter, health, mental health and social services. An additional component of this study will consist of a provider survey of health and other human service systems to gather information concerning identification of agencies serving homeless persons, numbers of homeless persons served, and linkages between system components. This data will be used to persons served, and linkages between system components. This data will be used to generate demand for services and flow of homeless between elements of the system matrix as well as to identify ways in which more effective linkages between agencies could be effected to deliver services to this population.